[kepler-dev] From <kar> to <kar:kar>

[Jing Tao]

ben, see my comments below the text.

Thanks,

Jing

Jing Tao
National Center for Ecological
Analysis and Synthesis (NCEAS)
735 State St. Suite 204
Santa Barbara, CA 93101

On Wed, 14 Apr 2010, ben leinfelder wrote:

> Jing,
> If we have a namespace, does Metacat try to find the schema so that it can 
> also validate it when the XML is uploaded?
> I know for MOMLs it was grabbing the DTD from a berkley server and doing some 
> validation based on that.
> If we go though the trouble of validating to a schema (with an established 
> namespace) we should probably have the schema available online for other 
> tools to use for their own validation needs (like Metacat).
>

Yeah, Metacat will need the namespace schema to validate the xml instance 
when the xml is uploaded. I am working on this now.


> Also, how much hardcoding of the namespace prefix was needed (like in 
> component search)? In theory you should be able to use any prefix you want 
> (kar:kar, bensKar:kar, 
> thisCrazyLongPrefixThatNoOneWouldEverUseInRealLife:kar) and as long as it 
> mapped to the correct full namespace you'd be fine. I know in practice this 
> can be difficult to do (especially when authoring XML with plain old 
> strings).

The prefix is not much hardcoded. I used the xpath like this:
/*[local-name() = 'kar']/karEntry. It just checks the local name.


>
> Thanks,
> -ben
>
> On Apr 14, 2010, at 4:32 PM, Jing Tao wrote:
>
>> This is a good idea. Since our kar schema used 
>> http://www.kepler-project.org/kar, I just adopted it.
>> 
>> I have changed the schema to 2.0.0 on both trunk and 2 branch.
>> 
>> Thanks,
>> 
>> Jing
>> 
>> Jing Tao
>> National Center for Ecological
>> Analysis and Synthesis (NCEAS)
>> 735 State St. Suite 204
>> Santa Barbara, CA 93101
>> 
>> On Wed, 14 Apr 2010, Matt Jones wrote:
>> 
>>> The namespace should include a version identifier so that we can
>>> differentiate the schemas as the kar format evolves over time.  So I 
>>> suggest
>>> the namespace be:
>>> 
>>> http://www.kepler-project.org/kar-2.0.0
>>> 
>>> (Let's reserve the 1.0.0 namespace for the old versions of kars that 
>>> shipped
>>> with Kepler 1.0.0).
>>> 
>>> Matt
>>> 
>>> On Wed, Apr 14, 2010 at 3:00 PM, Jing Tao <tao at nceas.ucsb.edu> wrote:
>>> 
>>>> hi, devs:
>>>> 
>>>> I just got a task today - enable the kar xml validation during the remote
>>>> component search.
>>>> 
>>>> I palyed around and found that we should put namespace definition on kar
>>>> xml instance in order to enable the validation.
>>>> 
>>>> Currently, the kar xml generated by kepler looks like:
>>>> <kar>
>>>> <karFileName> </karFileName>
>>>> <karFileSize> </karFileSize>
>>>> <mainAttributes> </mainAttributes>
>>>> <karEntry> </karEntry>
>>>> </kar>
>>>> 
>>>> I propose to change to:
>>>> 
>>>> <kar:kar xmlns:kar="http://www.kepler-project.org/kar">
>>>> <karFileName> </karFileName>
>>>> <karFileSize> </karFileSize>
>>>> <mainAttributes> </mainAttributes>
>>>> <karEntry> </karEntry>
>>>> </kar:kar>
>>>> 
>>>> After making above change and twist some other java code, the component
>>>> search with kar xml validation works.
>>>> 
>>>> I am not sure if the change from <kar> to <kar:kar> will cause any other
>>>> break in kepler. Just want to double check and let everyone know my 
>>>> change.
>>>> If you do see any issue, please let me know.
>>>> 
>>>> Thanks,
>>>> 
>>>> Jing
>>>> 
>>>> 
>>>> 
>>>> Jing Tao
>>>> National Center for Ecological
>>>> Analysis and Synthesis (NCEAS)
>>>> 735 State St. Suite 204
>>>> Santa Barbara, CA 93101
>>>> _______________________________________________
>>>> Kepler-dev mailing list
>>>> Kepler-dev at kepler-project.org
>>>> http://mercury.nceas.ucsb.edu/kepler/mailman/listinfo/kepler-dev
>>>> 
>>> 
>> _______________________________________________
>> Kepler-dev mailing list
>> Kepler-dev at kepler-project.org
>> http://mercury.nceas.ucsb.edu/kepler/mailman/listinfo/kepler-dev
>